Method and computer program product for estimating at least one state of a dynamic system

• US 6,711,476 B2
• Filed: 09/27/2001
• Issued: 03/23/2004
• Est. Priority Date: 09/27/2001
• Status: Expired due to Fees

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Speer, Thomas E., Jones, Richard D.
• Primary Examiner(s): Nguyen, Tan Q.
• Assistant Examiner(s): Tran, Dalena
• Application Number:
  US09/967,403
• Publication Number:
  US 20030060939A1
• Time in Patent Office: 908 Days
• Field of Search: 701/3, 701/4, 701/220, 701/226, 701/214, 701/207, 701/213, 701/221, 701/224, 701/5, 701/7, 701/8, 244/223, 244/180, 244/175, 244/181, 244/176, 244/191, 244/195, 244/178, 244/194, 244/192, 244/76.R, 244/196, 244/197, 318/628, 318/611, 700/29, 700/55, 700/46
• US Class Current: 701/3
• CPC Class Codes: G05B 13/04 involving the use of models or simulators

Abstract

A method and computer program product are provided for estimating the states of a dynamic system based upon a combination of first and second feedback signals. The first feedback signal is based upon anticipated state changes and is typically represented by a state propagation model. The second feedback signal is based upon the difference between the anticipated and actual measurements of the sensors. This difference may be weighted based upon a predetermined criteria, such as the credibility of the actual measurements and/or the degree to which outlying sensor measurements are to be discounted. The weighted difference is converted into a corresponding change in the state estimates. At least one feedback signal may then be modified to define its relative contribution to the state estimates. The first and second feedback signals are thereafter combined to produce the state estimates, subject generally to limitations upon the permissible changes in the state estimates.

Claims

1. A method of estimating at least one operational state of a dynamic system utilizing at least partially automated control and having at least one actuator and at least one sensor, the method comprising:

generating a first feedback signal at least partially based upon an anticipated change in a value representative of the at least one operational state of the dynamic system that is attributable to a current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator and independent of the actual measurement of any sensor;

generating a second feedback signal at least partially based upon a difference between a value of an anticipated measurement that is expected to be provided by the at least one sensor and a value of the actual measurement provided by the at least one sensor, wherein generating the second feedback signal comprises weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor based upon a predetermined criteria; and

combining the first and second feedback signals to produce an estimate of the at least one operational state of the dynamic system.

2. A method according to claim 1 wherein weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises separately weighting the difference between the respective values of the anticipated and actual measurements of each sensor based upon the relative credibility of the actual measurements provided by each respective sensor.

3. A method according to claim 1 wherein weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises weighting each difference with a predefined penalty factor having an effect that varies based upon the magnitude of a respective difference.

4. A method according to claim 1 further comprising limiting permissible changes in the estimate of the at least one operational state of the dynamic system based upon a predetermined criteria.

5. A method according to claim 4 wherein limiting the permissible changes in the estimate of the at least one operational state of the dynamic system comprises limiting the permissible rate of change of the estimate of the at least one operational state of the dynamic system.

6. A method according to claim 4 wherein limiting the permissible changes in the estimate of the at least one operational state of the dynamic system comprises limiting the estimate of the at least one operational state of the dynamic system to within a predefined range.

7. A method according to claim 1 further comprising modifying at least one of the first and second feedback signals prior to their combination in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the at least one operational state of the dynamic system.

8. A method according to claim 1 wherein generating the second feedback signal further comprises converting the weighted difference between the respective values of the anticipated and actual measurements of the at least one sensor into a corresponding change in the estimate of the at least one operational state of the dynamic system that is attributable to the weighted difference between the respective values of the anticipated and actual measurements of the at least one sensor.

9. A method according to claim 1 wherein generating the first feedback signal at least partially based upon the anticipated change in the value representative of the at least one operational state of the dynamic system comprises generating the first feedback signal at least partially based upon a state propagation model representative of the anticipated change in the value representative of the at least one operational state of the dynamic system based upon the current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator.

10. A method according to claim 1 wherein generating the second feedback signal at least partially based upon the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises generating the second feedback signal at least partially based upon a sensor output model representative of the anticipated measurement that is expected to be provided by the at least one sensor based upon the current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator.

11. A method according to claim 10 wherein generating the second feedback signal further comprises determining a first dot product of a vector representing the weighted differences between the respective values of the anticipated and actual measurements of each sensor and a transpose of a matrix representing changes in the sensor output model as a result of changes in the estimate of the value representative of the at least one operational state of the dynamic system.

12. A method according to claim 11 wherein generating the second feedback signal further comprises determining a second dot product of the first dot product and a gain matrix to thereby modify the second feedback signal relative to the first feedback signal in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the value representative of the at least one operational state of the dynamic system.

13. A method of estimating at least one operational state of a dynamic system utilizing at least partially automated control and having at least one actuator and at least one sensor, the method comprising:

generating a first feedback signal at least partially based upon an anticipated change in a value representative of the at least one operational state of the dynamic system that is attributable to a current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator and independent of the actual measurement of any sensor;

generating a second feedback signal at least partially based upon a value of an anticipated measurement that is expected to be provided by the at least one sensor and a value of the actual measurement provided by the at least one sensor;

combining the first and second feedback signals to produce an estimate of the at least one operational state of the dynamic system; and

limiting permissible changes in the estimate of the at least one operational state of the dynamic system based upon a predetermined criteria.

14. A method according to claim 13 wherein limiting the permissible changes in the estimate of the at least one operational state of the dynamic system comprises limiting the permissible rate of change of the estimate of the at least one operational state of the dynamic system.

15. A method according to claim 13 wherein limiting the permissible changes in the estimate of the at least one operational state of the dynamic system comprises limiting the estimate of the at least one operational state of the dynamic system to within a predefined range.

16. A method according to claim 13 wherein generating the second feedback signal comprises weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor based upon a predetermined criteria.

17. A method according to claim 16 wherein weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises separately weighting the difference between the respective values of the anticipated and actual measurements of each sensor based upon the relative credibility of the actual measurements provided by each respective sensor.

18. A method according to claim 16 wherein weighting the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises weighting each difference with a predefined penalty factor having an effect that varies based upon the magnitude of a respective difference.

19. A method according to claim 13 further comprising modifying at least one of the first and second feedback signals prior to their combination in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the at least one operational state of the dynamic system.

20. A method according to claim 13 wherein generating the second feedback signal further comprises converting the difference between the respective values of the anticipated and actual measurements of the at least one sensor into a corresponding change in the estimate of the at least one operational state of the dynamic system that is attributable to the difference between the respective values of the anticipated and actual measurements of the at least one sensor.

21. A method according to claim 13 wherein generating the first feedback signal at least partially based upon the anticipated change in the value representative of the at least one operational state of the dynamic system comprises generating the first feedback signal at least partially based upon a state propagation model representative of the anticipated change in the value representative of the at least one operational state of the dynamic system based upon the current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator.

22. A method according to claim 13 wherein generating the second feedback signal at least partially based upon the difference between the respective values of the anticipated and actual measurements of the at least one sensor comprises generating the second feedback signal at least partially based upon a sensor output model representative of the anticipated measurement that is expected to be provided by the at least one sensor based upon the current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator.

23. A method according to claim 22 wherein generating the second feedback signal further comprises determining a first dot product of a vector representing the weighted differences between the respective values of the anticipated and actual measurements of each sensor and a transpose of a matrix representing changes in the sensor output model as a result of changes in the estimate of the value representative of the at least one operational state of the dynamic system.

24. A method according to claim 23 wherein generating the second feedback signal further comprises determining a second dot product of the first dot product and a gain matrix to thereby modify the second feedback signal relative to the first feedback signal in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the value representative of the at least one operational state of the dynamic system.

25. A computer program product for estimating at least one operational state of a dynamic system utilizing at least partially automated control and having at least one actuator and at least one sensor, the computer program product comprising a computer-readable storage medium having computer-readable program code embodied in said medium, the computer-readable program code comprising:

a first executable portion adapted to generate a first feedback signal at least partially based upon an anticipated change in a value representative of the at least one operational state of the dynamic system that is attributable to a current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator and independent of the actual measurement of any sensor;

a second executable portion adapted to generate a second feedback signal at least partially based upon a difference between a value of an anticipated measurement that is expected to be provided by the at least one sensor and a value of the actual measurement provided by the at least one sensor, wherein said second executable portion is also adapted to weight the difference between the respective values of the anticipated and actual measurements of the at least one sensor based upon a predetermined criteria; and

a third executable portion adapted to combine the first and second feedback signals to produce an estimate of the at least one operational state of the dynamic system.

26. A computer program product according to claim 25 wherein said second executable portion is further adapted to separately weight the difference between the respective values of the anticipated and actual measurements of each sensor based upon the relative credibility of the actual measurements provided by each respective sensor.

27. A computer program product according to claim 25 wherein said second executable portion is further adapted to weight each difference with a predefined penalty factor having an effect that varies based upon the magnitude of a respective difference.

28. A computer program product according to claim 25 further comprising a fourth executable portion adapted to limit permissible changes in the estimate of the at least one operational state of the dynamic system based upon a predetermined criteria.

29. A computer program product according to claim 28 wherein said fourth executable portion is also adapted to limit the permissible rate of change of the estimate of the at least one operational state of the dynamic system.

30. A computer program product according to claim 28 wherein said fourth executable portion is also adapted to limit the estimate of the at least one operational state of the dynamic system to within a predefined range.

31. A computer program product according to claim 25 further comprising a fifth executable portion adapted to modify at least one of the first and second feedback signals prior to their combination in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the at least one operational state of the dynamic system.

32. A computer program product according to claim 25 wherein said second executable portion is further adapted to convert the weighted difference between the respective values of the anticipated and actual measurements of the at least one sensor into a corresponding change in the estimate of the at least one operational state of the dynamic system that is attributable to the weighted difference between the respective values of the anticipated and actual measurements of the at least one sensor.

33. A computer program product for estimating at least one operational state of a dynamic system utilizing at least partially automated control and having at least one actuator and at least one sensor, the computer program product comprising a computer-readable storage medium having computer-readable program code embodied in said medium, the computer-readable program code comprising:

a first executable portion adapted to generate a first feedback signal at least partially based upon an anticipated change in a value representative of the at least one operational state of the dynamic system that is attributable to a current estimate of the value representative of the at least one operational state of the dynamic system and the commanded state of the at least one actuator and independent of the actual measurement of any sensor;

a second executable portion adapted to generate a second feedback signal at least partially based upon a value of an anticipated measurement that is expected to be provided by the at least one sensor and a value of the actual measurement provided by the at least one sensor;

a third executable portion adapted to combine the first and second feedback signals to produce an estimate of the at least one operational state of the dynamic system; and

a fourth executable portion adapted to limit permissible changes in the estimate of the at least one operational state of the dynamic system based upon a predetermined criteria.

34. A computer program product according to claim 33 wherein said fourth executable portion is also adapted to limit the permissible rate of change of the estimate of the at least one operational state of the dynamic system.

35. A computer program product according to claim 33 wherein said fourth executable portion is also adapted to limit the estimate of the at least one operational state of the dynamic system to within a predefined range.

36. A computer program product according to claim 33 wherein said second executable portion is also adapted to weight the difference between the respective values of the anticipated and actual measurements of the at least one sensor based upon a predetermined criteria.

37. A computer program product according to claim 36 wherein said second executable portion is further adapted to separately weight the difference between the respective values of the anticipated and actual measurements of each sensor based upon the relative credibility of the actual measurements provided by each respective sensor.

38. A computer program product according to claim 36 wherein said second executable portion is further adapted to weight each difference with a predefined penalty factor having an effect that varies based upon the magnitude of a respective difference.

39. A computer program product according to claim 33 further comprising a fifth executable portion adapted to modify at least one of the first and second feedback signals prior to their combination in order to at least partially define the relative contributions of the first and second feedback signals to the estimate of the at least one operational state of the dynamic system.

40. A computer program product according to claim 33 wherein said second executable portion is also adapted to convert the difference between the respective values of the anticipated and actual measurements of the at least one sensor into a corresponding change in the estimate of the at least one operational state of the dynamic system that is attributable to the difference between the respective values of the anticipated and actual measurements of the at least one sensor.